* transformers 4.47.1 * drop monkeypatches * can't remove patches yet * make flash attention forward ignore the loss kwargs * patch the flash attention in the modeling arch too * remove fsdp and deepspeed patches * cleanup PR * bump accelerate and torchao, also logically reorder/group requirements * meant to include torchao * use official patch release
37 lines
1.0 KiB
Python
37 lines
1.0 KiB
Python
# noqa
|
|
# pylint: skip-file
|
|
try:
|
|
import torch
|
|
except ImportError:
|
|
raise ImportError("Install torch via `pip install torch`")
|
|
from packaging.version import Version as V
|
|
|
|
v = V(torch.__version__)
|
|
cuda = str(torch.version.cuda)
|
|
try:
|
|
is_ampere = torch.cuda.get_device_capability()[0] >= 8
|
|
except RuntimeError:
|
|
is_ampere = False
|
|
if cuda != "12.1" and cuda != "11.8" and cuda != "12.4":
|
|
raise RuntimeError(f"CUDA = {cuda} not supported!")
|
|
if v <= V("2.1.0"):
|
|
raise RuntimeError(f"Torch = {v} too old!")
|
|
elif v <= V("2.1.1"):
|
|
x = "cu{}{}-torch211"
|
|
elif v <= V("2.1.2"):
|
|
x = "cu{}{}-torch212"
|
|
elif v < V("2.3.0"):
|
|
x = "cu{}{}-torch220"
|
|
elif v < V("2.4.0"):
|
|
x = "cu{}{}-torch230"
|
|
elif v < V("2.5.0"):
|
|
x = "cu{}{}-torch240"
|
|
elif v < V("2.6.0"):
|
|
x = "cu{}{}-torch250"
|
|
else:
|
|
raise RuntimeError(f"Torch = {v} too new!")
|
|
x = x.format(cuda.replace(".", ""), "-ampere" if is_ampere else "")
|
|
print(
|
|
f'pip install unsloth-zoo==2024.12.1 && pip install --no-deps "unsloth[{x}]==2024.12.4"'
|
|
)
|